The Road to the 50% Thermally Efficient F1 Internal Combustion Engine

[johnny comelately]

[Tommy Cookers]

yes 52% thermodynamic efficiency is remarkable

how does this stand ? ......
at partial power
with exhaust catalysis
at higher lambda

iirc we have natural gas ICEs running over 2 lambda ie 'heat dilution' - unlike F1 MGU-H there's no compounding
does the F1 coating so reduce heat dumped to coolant as to make such high lambda unworthwhile ?

does lean NOx catalysis need heat ?
the stoichiometric-mass heat of natural gas might be a bit higher than F1 gasoline's
does a NOx catalyst need to be upstream of the MGU-H turbine ?

etc

[johnny comelately]

yes 52% thermodynamic efficiency is remarkable

how does this stand ? ......
at partial power
with exhaust catalysis
at higher lambda

iirc we have natural gas ICEs running over 2 lambda ie 'heat dilution' - unlike F1 MGU-H there's no compounding
does the F1 coating so reduce heat dumped to coolant as to make such high lambda unworthwhile ?

does lean NOx catalysis need heat ?
the stoichiometric-mass heat of natural gas might be a bit higher than F1 gasoline's
does a NOx catalyst need to be upstream of the MGU-H turbine ?

etc

From the lean of peak temperatures of 1.4 lamda produce less NOx, I think Pat Symonds mentions this, could be wrong.
The coatings have to be there because of the reduced or no boundary layer
To enable no quench they have to be a hotter than typical surface, is this shown by the 3.5 bar coolant temperature produced?
imho

[vorticism]

https://images-wixmp-ed30a86b8c4ca88777 ... 4ofgyvuIWM

Yes, this is what some here are postulating. The CC roof won't have that much curvature, imo, more flat like a diesel. This is borne out by the camshaft spacing visible on the timing covers.

[johnny comelately]

https://images-wixmp-ed30a86b8c4ca88777 ... 4ofgyvuIWM

Yes, this is what some here are postulating. The CC roof won't have that much curvature, imo, more flat like a diesel. This is borne out by the camshaft spacing visible on the timing covers.

The curvature was for illustrative purposes only.
Bear in mind that the more spherical (curved) it is the lower the surface area which relates to lost heat. The tiny chamber volume is a challenge. Any corners or similar anomalies are an obstacle for full burn and speed and potential knock origin, normally. But the use of HCCI with spark augmentation would get rid of those problems, I was thinking about this when considering the omega forms.
In reality it could be imagined the prechamber could be canted and some of the holes to be somewhat tangental.
Exploration of the characteristics of the HCCI affect is a complete chapter and ideas and discussion about it would be welcome

[gruntguru]

yes 52% thermodynamic efficiency is remarkable

how does this stand ? ......
at partial power
with exhaust catalysis
at higher lambda

iirc we have natural gas ICEs running over 2 lambda ie 'heat dilution' - unlike F1 MGU-H there's no compounding
does the F1 coating so reduce heat dumped to coolant as to make such high lambda unworthwhile ?

does lean NOx catalysis need heat ?
the stoichiometric-mass heat of natural gas might be a bit higher than F1 gasoline's
does a NOx catalyst need to be upstream of the MGU-H turbine ?

Of course the whole engine must be designed for peak TE at peak power. If the goal was peak TE only, it might be achieved at some lower load point and be even higher than 52%.

The requirement for aftertreatment would be reduced. Anecdotally the current trend to downsized, highly turbocharged engines in road cars has spawned vehicles which are gross emitters at high load (rich AFR to curb knock and thermal loads). High loads are not required during emissions test drive cycles, however many of these vehicles are frequently operated at high loads to enjoy the available performance.

I believe the current F1 engine is not operated at higher lambda because of one or all of the following:
- mechanical limits at 400 bar peak cylinder pressure
- charging pressure ratios already at the efficiency limits for single stage turbomachinery (multi-stage with inter-stage cooling would be a game changer. Road applications would not require such high PR due to lower specific output.)
- pre-chamber / main chamber mixture requirements difficult to control with single DI injector

[vorticism]

Anecdotally the current trend to downsized, highly turbocharged engines in road cars has spawned vehicles which are gross emitters at high load (rich AFR to curb knock and thermal loads). High loads are not required during emissions test drive cycles, however many of these vehicles are frequently operated at high loads to enjoy the available performance.

It's too bad the corrupt gov's of the world held VW to such account for doing the same thing; not honoring the spirit of their rules, only their letter. One wonders what could have been of more R18s.

Would a diesel in F1 look anything like the current V6s? In some ways you might say yes: DI & TC, high compression. However the properties of the fuels must be significant; yet are the differences between CI and HCCI so great?

[gruntguru]

However the properties of the fuels must be significant; yet are the differences between CI and HCCI so great?

There is a huge difference between premixed (typical of SI) and diffusion (typical of diesel) combustion.

HCCI is a premixed combustion regime.

It might be possible to approach the F1 combustion regime in a diesel with a form of PPCI (Partially Premixed Compression Ignition) where some of the charge is pre-mixed e.g. an early spray of fuel near BDC and the remainder of the fuel is injected near TDC to initiate combustion. This technology is fairly mature in the lab. Not sure if researchers have yet achieved HCCI using this technique.

[vorticism]

Is it correct to say that the fuel injection or mixing is what limits the cycle speed of an optimized racing diesel engine? The Audi LMP1 diesels in their top form V6s redlined around 6k iirc. Would a diesel fuel ever be able to run in higher engine speeds? Maybe the method you mentioned get it some of the way there. Or is it relegated to making more power at lower engine speeds with the drawback of requiring heavier engine components.

[johnny comelately]

Is it correct to say that the fuel injection or mixing is what limits the cycle speed of an optimized racing diesel engine? The Audi LMP1 diesels in their top form V6s redlined around 6k iirc. Would a diesel fuel ever be able to run in higher engine speeds? Maybe the method you mentioned get it some of the way there. Or is it relegated to making more power at lower engine speeds with the drawback of requiring heavier engine components.

If everything hangs together the limiting factor is usually air supply (I know, highly arguable)
I always viewed the current F1 engine at say 11K RPM was about right for the turbo design, no need to rev higher (BMEP). But with more consideration, the relatively enormous pressures involved must call for heavier components which limit RPM. So, all in all the format works.

[gruntguru]

Yes, but vorticism is right. Diesel speed is limited by the time required to get every fuel molecule into contact with enough oxygen molecules to assure complete combustion. The fuel molecule at the centre of a droplet of diesel spray has more than a hundred layers of fuel molecules between it and the air - for even the smallest droplet of injected fuel.

This is the big disadvantage of diffusion combustion over premixed combustion.

[Tommy Cookers]

... The fuel molecule at the centre of a droplet of diesel spray has more than a hundred layers of fuel molecules between it and the air - for even the smallest droplet of injected fuel.

c.120 years ago Ackroyd Stuart's CI engine using air-blast fuelling would have alleviated this situation ?

(but the backers of Diesel's inferior system 'stole in the courts' the CI engine - and called it the Diesel engine)

https://tractors.fandom.com/wiki/Herbert_Akroyd_Stuart

[vorticism]

As it relates, the first several seasons of this formula had its oil burning phenomenon. Was it ever determined if the oil was injected port side or directly? As liquid or aerosol? As a vapor component off gassing? For the purposes of reacting with the main fuel (RCCI?), or as additional fuel energy content?

[Zynerji]

However the properties of the fuels must be significant; yet are the differences between CI and HCCI so great?

There is a huge difference between premixed (typical of SI) and diffusion (typical of diesel) combustion.

HCCI is a premixed combustion regime.

It might be possible to approach the F1 combustion regime in a diesel with a form of PPCI (Partially Premixed Compression Ignition) where some of the charge is pre-mixed e.g. an early spray of fuel near BDC and the remainder of the fuel is injected near TDC to initiate combustion. This technology is fairly mature in the lab. Not sure if researchers have yet achieved HCCI using this technique.

I always thought that was exactly how the single injector TJI system worked...

[Airshifter]

Is it correct to say that the fuel injection or mixing is what limits the cycle speed of an optimized racing diesel engine? The Audi LMP1 diesels in their top form V6s redlined around 6k iirc. Would a diesel fuel ever be able to run in higher engine speeds? Maybe the method you mentioned get it some of the way there. Or is it relegated to making more power at lower engine speeds with the drawback of requiring heavier engine components.

If everything hangs together the limiting factor is usually air supply (I know, highly arguable)
I always viewed the current F1 engine at say 11K RPM was about right for the turbo design, no need to rev higher (BMEP). But with more consideration, the relatively enormous pressures involved must call for heavier components which limit RPM. So, all in all the format works.

NA or boosted, the rev range, torque curves, weight considerations, etc will all play a part in component weights. Every change that might seem minor to us could be a huge design compromise at some point.

I was actually surprised that cars were approaching 11K. Granted they want revs to produce the horsepower, but the instantaneous acceleration of pistons and such causes a lot of over engineering with increased revs. I would have thought the payoff would reach a decreasing scale at lower revs.

Once this type of level is reached, things are going to get complicated quickly in terms of improvements. Very small gains might take big packaging changes. As an example, with all the new injection/ignition thoughts and development taking place simply increasing dwell time might easily make things more precise and easier to control. But with already long (relative) rod ratios, that's a lot of heavy components to alter to increase the rod ratio even more.